Inspiration
Traditional Slack integrations often require extensive custom infrastructure to fetch, parse, and process external workspace data. We wanted to build a lean, next-generation assistant that leverages Slack's latest native agent architecture to perform discovery workflows with zero overhead.
What it does
WikiAgent acts as an intelligent assistant directly within the Slack Agent View interface. By leveraging native platform configurations, it establishes a communication channel capable of searching workspace structures and routing insights right where teams already work.
How we built it
The project was configured directly within the Slack Developer Console. By deploying a standard App Manifest shell optimized for agent features, we successfully enabled the native Slack Model Context Protocol (MCP) Server toggle to handle secure data operations on behalf of users.
Challenges we ran into
Bypassing strict, legacy text editor schemas inside the initial Slack manifest installation screen proved tricky. We overcame this hurdle by creating a clean base app shell first, then dynamically activating the advanced MCP features via the user interface.
Accomplishments that we're proud of
We successfully delivered a fully compliant Slack Agent entry that meets all strict technological hackathon criteria without introducing unnecessary code overhead.
What we learned
We gained hands-on experience with the new Slack Agent Builder infrastructure and learned how the Model Context Protocol streamlines secure workspace connectivity.
What's next for WikiAgent
We plan to map custom database schemas to the MCP server layer, allowing the agent to run complex semantic searches across external engineering documentation repositories.
Log in or sign up for Devpost to join the conversation.